Thoughts on Chart Pacing

[Milliways]

Since reviving my personal chart a couple of weeks ago, I’ve been giving some consideration to its pacing. That is, how high new songs enter and how quickly they leave the chart.

In previous incarnations of the chart, I’ve weighted it heavily towards the newest songs, such that new entries high up, even at #1, are common and runs in the chart tend to be short. This has come about to a large extent because I’ve tended to concentrate my listening on each week’s new releases playlist, rather than on the existing chart.

But, it has a few disadvantages – not least, that it’s out of step with the way other Haveners compile their charts, which means that a lot of my hits are unlikely to contribute to the Combi Chart, since they’ve peaked and fallen down or out of my chart before they climb in other charts.

So, I’m thinking that this time around I will make more effort to listen to the songs in the chart on more of an ongoing basis, rather than focusing so much on the newest releases, and run a slower chart. I’m
hoping I’ll get enough listening time each week to make it work, even as work gets inevitably more hectic!

[TheThorne]

Well having run a chart non stop for 33 years, I always find the pace themselves. I remove all the leavers and add new songs into the spaces which can be 14-22 new entries. Then I start from the bottom and work my way up. This can make it hard for songs low down to climb as they keep getting pushed back down by exciting new songs but it also means songs entering within my top 30 are very rare, usually only massive returning bands get that privilege. All songs have to climb but entering above 60 ensures they will have success.

As for listening I have 2 playlists, the current chart and new entries and I tend to play the current chart more so another reason why they enter lower but I also hear most off my new entries on the radio and if I hear them a lot, I heard the new 1975 6 times in one day, that obviously will boost their chances. I also tend to midweek and add songs in but my chart day I have more I need to add and again that pushes out low songs.

[Razzle Dazzle]

The size of a chart has a huge effect on the pacing. When I started with a top 30 great songs went straight in the top 5, good songs in the top 15 and the rest made up my top 30 so there was naturally a quick turnover. Over the years I extended it to a top 100 and found the same songs that would start a lot lower and have to rise.

I like to give good songs about 8 or 9 weeks in my chart so a chart run of a #1 song is usually something like 43/31/22/10/4/1/7/23/67.

Reasons I guess, first i use my chart as a playlist so i want my 20 favourite songs of the month to be my top 10 for as long as possible.

[frag]

My chart's pace has gotten slower as I've gotten older. In my initial run (1999-2004, top 10/20/30), most #1s would be new entries or non-movers, with very few risers in the charts at all. I had more time and fewer songs to listen to, and the songs I did listen to were more popular so they'd get radio play or discussion at school.

When I started again in 2005 through to 2011 (with a top 40 and maintaining a top 75 below that), it became less common for songs to enter in the top 10, although there would usually be an entry around 11-15. I still found it strange when I saw charts like TheThorne's where songs would get entered weeks in advance and take a month or more to reach #1.

Nowadays, it's rare for songs to enter my top 20 (and new songs entered the top 10 only twice this year), and it's getting more common for songs to come from outside the top 40 and into the top 20. I don't think this is on purpose, but I do think it's self-perpetuating: songs from this week have to queue up a bit below songs from previous weeks that are on their way up. Also, usually I've only listened to a song two and a half times before charting it (half because my weekly longlist is too long to hear each song in full, so I just listen long enough to decide whether to playlist it properly), so this week's ones usually haven't had time to get in my head yet.

[o]

Speaking from a combi chart compiler point of view, I find it annoying that certain songs seem to fall out of a top 30 to outside it in a week in some of the above charts. I might see a song in your charts, listen, like it, chart it, and then the following week, one of both of you kick it back out to 37. So from a combi chart perspective, it can be frustrating for someone that wants to see more indie songs in that chart, but these are your own charts, so run them how you like I'm struggling to get 30 songs in mine, so heck knows how you have 100 in yours, I dont have the time to listen to that amount of songs, let alone try and arrange them 1-100 each week, so fair play to you!

[TheThorne]

I don’t do combichart as I have to recompile it as my band and title are wrong way round, they come straight that way from Spotify

[o]

I add your chart in every week, and add them all in manually

[Razzle Dazzle]

Most of my songs in my top 20 hang around for 2 month, I'm still clearing out November songs, the problem is in that time to keep a top 100 chart moving it is going to see 120/150 new entries (about 15 a week) and if my top songs stick around and the bottom songs climb the only place for the middle is to quickly fall away

[Haguey]

I've changed it several times over the years but I have a 'system' now where everything gets a rolling 4-week 'score' so everything get's smoothed out and gives me more accurate, realistic chart runs.

It does give me a few odd climbs to number one but I like a chart run to look 'natural'.

53-45-21-5-3-1-1-2-2-3-5-7-11-15-21-29-36- was a recent chart run which is just about perfect for me for a big hit.

[frag]

Feb 1, 2020 20:25:20 GMT 1 TheThorne said:

I don’t do combichart as I have to recompile it as my band and title are wrong way round, they come straight that way from Spotify

TheThorne how do you get your chart from Spotify? I have a little Java program that I run every week. At some point someone complained that they're the wrong way round, so I started printing them the right way around.

That code's not very user-friendly at the moment, but I'd be happy to make it so  

[frag]

Feb 1, 2020 20:19:57 GMT 1 o said:

Speaking from a combi chart compiler point of view, I find it annoying that certain songs seem to fall out of a top 30 to outside it in a week in some of the above charts. I might see a song in your charts, listen, like it, chart it, and then the following week, one of both of you kick it back out to 37. So from a combi chart perspective, it can be frustrating for someone that wants to see more indie songs in that chart, but these are your own charts, so run them how you like I'm struggling to get 30 songs in mine, so heck knows how you have 100 in yours, I dont have the time to listen to that amount of songs, let alone try and arrange them 1-100 each week, so fair play to you!

That can be frustrating, yeah, but there's so much new stuff out there - I've been starting with 90 songs each week lately, and this week have a playlist of 34 songs. There's usually about 4-6 songs going into and out of my top 30 each week, and more often than not they'll rise from outside rather then new songs going straight in there.

[TheThorne]

Feb 4, 2020 23:08:57 GMT 1 frag said:

Feb 1, 2020 20:25:20 GMT 1 TheThorne said:

I don’t do combichart as I have to recompile it as my band and title are wrong way round, they come straight that way from Spotify

TheThorne how do you get your chart from Spotify? I have a little Java program that I run every week. At some point someone complained that they're the wrong way round, so I started printing them the right way around.

That code's not very user-friendly at the moment, but I'd be happy to make it so  

Oh it’s nothing sophisticated, I just copy and paste 

[Jordan]

I run a top 100 and songs very rarely debut in the top 50. Those that do have a tendency to reach the top. I have a few playlists that I use so I play older songs as often as I do newer ones, possible more so.

[Earl Purple]

If songs peak on the 2nd / 3rd chart week then you have far more chance to "change your mind" about what was the song of the week, and in my final year of charts, 2018, several singles got to #1 that were not the highest new entry on their week. I have had occasions (not so many in 2018) where a single that got to #1 was way down the list of new entries.

One thing that shaped the last 4 years in my chart to give me slow climbers was the video show. Before then it was the chart countdown that sometimes led to these, i.e. every week I would listen to a countdown of my existing chart.

Retro charts suffer though from me scoring up a playlist then abandoning it as I move on. So whilst I have "Life On Mars" as my #1 in 1973 I'm not listening to that song right now - instead I have a playlist that has The Laughing Gnome on it - and actually quite a number of classics. I will only revisit 1973's big hits when I do the chart of the year. And what a year it was!

[Milliways]

I’ve commented on this week’s chart that it has been an example of a new releases playlist that had one very immediate track that eclipsed all the others. Despite this, I’ve only entered it at #10, whereas in previous iterations of the chart, it would have been straight in at #1 and Best Coast (the actual #1, which has climbed for 2 weeks to get there) would have been out of luck. As you say Earl Purple, it means I have the chance to change my mind about it, if I go off it sooner than I expect then it won’t make the top spot.

One aspect of my chart construction, still, that I appreciate isn’t ideal from a Combi Chart point of view is how quickly some of the older hits are now falling out (I’ve noted o’s post above). Radkey is a case in point, it’s peaking in the current Combi just as it’s had its time at the top in mine and is on its way down after 5 weeks on the top spot. A Top 40 is the longest weekly chart I’ve ever done, and any longer is just going to be too burdensome to do on a weekly basis especially when work gets more hectic. I don’t know how some of you who do top 100s manage it, especially if participating in other Haven events (Dance Idol, Havenvision etc) at the same time. Like Razzle Dazzle, I do want to keep the chart moving and be able to fit in new entries to then listen to over the following couple of weeks, and the only way I can do that is by having old songs fall out quickly when they’ve started to go stale. Realistically, only the very strongest songs are going to manage much more than a 5-week run.

[TheThorne]

I spent about 2 hours on my chart in total and that includes making new release playlist and adding to it through the week, the actual chart only takes me about 15 mins.

When you have been doing it 34 years non stop it’s just simple, I blank out the songs that are leaving and add my favourite new ones and before I publish my chart a few midweek releases can sneak in and push out others, I start from 100 and give each song a number when I get to 1, I shuffle it into the correct order. I tried to stick to rules like top 10 songs don’t drop out of the top 25 and top 40 songs won’t drop out of the 75 although there is that odd exception if I have too many grats, I had 4 Bombay Bicycle Club songs in my chart at once.

[Razzle Dazzle]

I just copy and paste my previous chart over

Update my spotify playlist with new releases and the order I want them in

And just reorder them adding the new releases in, its slot of copy and pasting buy usually takes half an hour, I think I would prefer a top 75 because 100 is too big but I actually listen to my songs that I chart 100-75 in my playlist and good songs do enter down there and rise to be my favourite songs of the year so I dont mind taking an extra 10 minutes making a top 100.

[Earl Purple]

I have a formula that generates the positions and all I have to do is rank the new singles. I twiddle a few and the rest just follow the formula.

With a top 100 it came down eventually to approximately 18 new singles a week. It's always been close one fifth of a chart being new entries regardless of the chart size.

It was truly burdensome, firstly creating the playlist albeit I had programs and formulae to help find the initial list, and then ranking them. There were times I gave songs points and any tied were just randomly ordered. The top ones always still came out on top, but given that there were so many similar-sounding kind of songs it was difficult.

[raliverpool]

I do what I've always done, Assemble my favourite new(ish) track at that day as my #1 song whether that is a 5 week climber or a brand new Friday release debuting at the summit; and reassemble my previous week chart accordingly to make a new chart.

What I do throughout the week is add new songs to my new Spotify music playlists from certain music blogs (Pitchfork; Stereogum; Set Music Free; Popjustice; Rolling Stone; NME); plus youtube's New Alternative 40 chart .... then on Friday see if I've missed any gems added to Spotify's New Music Friday UK & Release Radar playlists; and new tracks added to BBC Radio 1, 2 & 6Music Playlists (thanks Blondini for now adding the videos saving me some time & effort) which I've missed & reconfigure ...

and voila I have a new Top 50 chart.

My turnover seems to be circa 12 new entries a week.

[djshaun]

My approach is pretty simple which is to review the new releases on a Friday plus any midweek releases that have sneaked in and then update my chart accordingly, taking out those I have become bored with. This does mean my chart fluctuates from a top 40 generally, down to a top 20 briefly at one point last year when there was a real glut of new music that I liked...and the highest I got up to I think was a top 75 again very briefly I think about 5 years ago. In terms of rankings there’s no magic formula, it is just based on my preference at the time although when I listen to my chart...usually driving to and from work, I am thinking about whether each song is better than the next (which it should be if I’ve got it right!) but that also then gives me a sense of whether a song will be climbing the next week or dropping big time if I’ve become fed up with it.

I do get bored listening to the same songs over and over so I like to keep mine fresh and it is unusual for a song to stay in my chart for more than 8 or 9 weeks. In addition, I only tend to listen to my chart once a week, usually at the start of the week, and then the rest of the week I rotate my playlists through my old charts of the year. Since I stopped dj’ing I’ve had more time on my hands and have now created a chart of the year for every year from 1970 to date...the older years have as little as 30 songs but then some of the early 10s have as many as 350 in my chart of the year...I also tend to keep it varied by going from 70s to 90s to 10s to 80s to 00s so I won’t hear the same playlist again for potentially a couple of years!

It’s probably way too organised but it keeps me happy lol!